rename in-memory channel#700
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: sbezverk If they are not already assigned, you can assign the PR to them by writing The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
/assign @Harwayne |
Signed-off-by: Serguei Bezverkhi <sbezverk@cisco.com>
Signed-off-by: Serguei Bezverkhi <sbezverk@cisco.com>
|
/assign @grantr for approval |
|
@sbezverk: GitHub didn't allow me to assign the following users: for, approval. Note that only knative members and repo collaborators can be assigned. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
Signed-off-by: Serguei Bezverkhi <sbezverk@cisco.com>
|
What happens to existing channels using the |
|
@Harwayne the idea is, as long as CCPs are using old config map name, it should be transparent. I did not find any other dependencies. |
|
Thanks @sbezverk!
IIUC, after the upgrade there will be two CCPs each with an identical set of in-memory channel infrastructure. They'll both work, but future upgrades will only affect the new one. In a later release we'd want to delete the old one somehow. This PR renames a bunch of objects that the operator and developer rarely/never need to know about: Would the upgrade be easier if only the CCP object was renamed? We could leave the service accounts, deployments, etc that already exist in place, create a new CCP with the new name, and update the controller/dispatcher to handle both CCP names for this release. /hold @Harwayne just mentioned that he has more thoughts on this PR, so I'm holding it to give him time to weigh in. |
|
Talked with @grantr and have been convinced that we should try to re-use the existing infrastructure ( |
|
@grantr small correction. I think we should go with a separate controllers, one in-memory-channel (old) and in-memory controller with async (new) running in parallel for backward compatibility. |
|
I think this was superseded by #708. If so, can this PR be closed? |
|
@Harwayne @grantr it's now https://github.com/knative/eventing/tree/master/cmd/in_memory I guess... we are good? |
|
We have moved to in-memory and now are moving to CRDs over provisioners. |
|
@n3wscott: Closed this PR. DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
Signed-off-by: Serguei Bezverkhi sbezverk@cisco.com
Fixes #676